WHAMMY DIY Build

[Sythrix]

Yes, as I noted in another thread, volume control position can make significant difference in performance of the amplifier. I can make such measurements but there are infinite possibilities so not sure how to make comparisons between two devices.

Seems like the only thing you could do is min/max the volume and note the differences in power output alongside the measurements.

I was a little awestruck with the amount of information he was giving me and just how many ways he could spin it. It's funny, because if you were a manufacturer marketing an amp, you would have so many options... there needs to be a standard that everyone can conform to.

At least common ground can be found on most DACs.

EDIT: Or if you had a series of microphones in a controlled setting, you could reach equivalent dBs with various amps?

 

[JoelN]

I'm planning on building the WHAMMY headphone amplifier and have been going over the guide and schematic provided. It states the following if you are going to use the "LED reference" configuration" as you did: "If using D5 and D6 LEDs (as you are) omit R9 and R13. R9 and R13 may be shorted for +-15. I notice your build has resistors in the R9 and R13 position. Are you sure that is correct? The example LED reference example build on the DIYAudio site does appear to omit these resistors. I believe R10 and R14 are also omitted.

 

[Sythrix]

I'm planning on building the WHAMMY headphone amplifier and have been going over the guide and schematic provided. It states the following if you are going to use the "LED reference" configuration" as you did: "If using D5 and D6 LEDs (as you are) omit R9 and R13. R9 and R13 may be shorted for +-15. I notice your build has resistors in the R9 and R13 position. Are you sure that is correct? The example LED reference example build on the DIYAudio site does appear to omit these resistors. I believe R10 and R14 are also omitted.

No, actually. I'm not sure. It's working perfectly though, so I'm not clear on the implications. It looks like the power supply test was fine, but then I most likely got carried away when putting in resistors and didn't realize I had filled them.

Probably something to ask in the official thread, as they would be able to speak of the ramifications.

 

[Sythrix]

After looking over the schematic, I'm going to go back and cut out the resistors and see what happens. I don't know how I managed to overlook that, but I imagine it working OK without any obvious errors had something to do with it.

I even had it measured by AP at RMAF and they said everything looked good electrically.

Thanks for pointing it out though. I'm really annoyed with myself for letting it happen.

 

[restorer-john]

Nicely documented build with great pictures.

 

[Sal1950]

Thanks for pointing it out though. I'm really annoyed with myself for letting it happen.

Not many who could do near as good, let alone any better.

 

[Sythrix]

OK. I made the modification and took out the 4 resistors... I can't tell any difference .

However, it makes me feel better it's now at the purported lowest noise spec. Not being an electrical engineer, I don't know the ramifications, but maybe someone could tell us sometime what happened. Perhaps it defaulted to the "resistor reference" configuration?

I have no way of measuring it again to find out if there were any changes there. My ears can't tell the difference though .

Thanks again for pointing it out JoelN.

 

[Johnb]

Back again, after a longer hiatus than I had anticipated. Getting the drilling done was easy enough, but I lost a lot of motivation afterwards and just didn't feel like working on it. Fast forward to today and I'm only a few steps away from the testing stage.

View attachment 16107
Me and my uncle got the plates all drilled out. It's not quite perfect, but it's good enough for DIY. We accidentally went one size too large on the stepped bit for the headphone jack, but it didn't turn out to be a big deal.

View attachment 16108
At first, attacking the shielding to separate it was a real chore. I didn't understand quite how to do it and started with a flat-head screwdriver, as I has seen others do. Turns out the tip of an old multi-meter probe was a far better tool. You can actually unravel them if you take it in steps and have a precise tool.

View attachment 16109

View attachment 16110

View attachment 16111

View attachment 16112

View attachment 16113
After soldering both the signal wire and the shield, I heat-shrunk them all closed and did the other side in red.

View attachment 16114
I continued the process for the headphone jack.

View attachment 16115

View attachment 16116
I asked in the official WHAMMY forums on diyAudio and was told that the capacitor that is shunted from the RCAs to ground is a non-critical value, so I used an old one, which had previously turned out to be too big for one of my other projects. I'm hoping I won't have any issues.

View attachment 16117
Creating the threads with these screws turned out to be a much bigger chore than I had anticipated. The screws themselves kept wanting to strip and the aluminum on this panel proved to be far more formidable that it first appeared, making turns difficult.

View attachment 16118
In order to fit everything and have it away from the transformer on the PCB, we had to stack the RCAs like this in order to have the IEC Inlet in an appropriate position.

View attachment 16119
Back panel complete.

View attachment 16120
Had to widen the screw holes on the front panel headphone jack and use a nut on the inside in order get it straight. Unfortunately, the initial measurements were slightly crooked, so it required tweaking.

View attachment 16121
...and that's it for now. Next time I will probably finish. I'm going to include some thermal shots of the amp in action after it has warmed up (providing it works, of course), so that should be interesting.

I had been eyeing that one for a while, although the build instructions were non-specific for a number of choices to be made. That makes comparing results with other builders difficult. I used to love soldering through-hole components. Can't wait to hear what you make of the sound quality. And kudos for the effort.

 

[Sythrix]

I had been eyeing that one for a while, although the build instructions were non-specific for a number of choices to be made. That makes comparing results with other builders difficult. I used to love soldering through-hole components. Can't wait to hear what you make of the sound quality. And kudos for the effort.

It was a great experience and I'd recommend it.

As I said in my initial post, the directions can be a bit barebone in some places... glossing over what I would consider key details... but it's by no means a huge barrier. Just takes a little more research.

I'm not one to go on and on about sound quality. I'll say it seems fairly neutral and transparent to me, with a healthy amount of wiggle room, should you get anything demanding. I stop at 9-10 o clock with my dt 1990s (250 ohm) and it's painful at 12.

Not an amp for IEMs though, as with sensitive ones I can hear a moderate amount of hiss (versus nothing with my 1990s).

 

[JoelN]

I have no way of measuring it again to find out if there were any changes there. My ears can't tell the difference though .

Thanks again for pointing it out JoelN.

I'm glad you got things squared away and it still sounds OK. Who knows, keeping the wrong configuration may have had some unwanted consequences in the long run. I enjoyed reading about your build. Thanks for posting it. It will be very useful to me as I build my own WHAMMY.

 

[Sythrix]

Who knows, keeping the wrong configuration may have had some unwanted consequences in the long run.

Yes, that may be. Better it’s been dealt with.

I enjoyed reading about your build. Thanks for posting it. It will be very useful to me as I build my own WHAMMY.

Glad to hear it and good luck with your build!

 

[Sal1950]

Not being an electrical engineer, I don't know the ramifications, but maybe someone could tell us sometime what happened.

Now with the resistance removed, the signal can travel thru the circuit faster. I do struggle sometimes with the modern audiophool vocabulary of flowery descriptors, but sound will be significantly quicker and more musically free flowing.

 

[Sythrix]

Now with the resistance removed, the signal can travel thru the circuit faster. I do struggle sometimes with the modern audiophool vocabulary of flowery descriptors, but sound will be significantly quicker and more musically free flowing.

Ah... ah yes! Of course! How silly of me. Thank you for those... insightful... clarifications.

 

[Sythrix]

An update. After removing the unnecessary resistors, I discovered something. If I max the volume with no sound playing, I can now hear virtually nothing (maybe a distant, faint hiss?), whereas before when I maxed the volume, I would hear hissing and faint interference noise. There's definitely some kind of improvement that happened from removing them.

Now, I haven't had to go anywhere near max volume, but it's something worth note, just in case I ever use a more demanding headphone on this amp.

 

[trl]

Resistors R9, R10, R13 and R14 have nothing to do with the output background noise. You probably did something else if now with the same IEMs the background noise got lowered. I suggest you to swap OPA2604 with OPA1612 or OPA1652 (if you really want FET-input opamp) and see if noise gets lowered. Of course, LM4562/LME49720 mioght be tested as well.

Also, lowering the gain by adjusting the resisitve divider R8/R12 could help too from a gain of +14dB (5X?) to something lower, though existing gain seems well-chosen.

Measurements for power rails and also for the output audio signal are very good: https://www.diyaudio.com/forums/pas...pass-diy-headphone-amp-guide.html#post5317272. Seems that indeed the C-R-C filter done by the 5.1 Ohms and 3300uF caps is very efficient.

Also, schematics from DIY-Audio here:

 

[Sythrix]

Resistors R9, R10, R13 and R14 have nothing to do with the output background noise.

Thanks for clarifying that.

You probably did something else if now with the same IEMs the background noise got lowered.

It's actually my DT 1990s where I noticed the change with max volume. It makes no difference in the normal range of volume. When I maxed the volume before, there was definitely some noticeable noise and hiss, whereas now there is nothing. Both times the amp was connected to my computer and on, but with the sound muted from the computer's output. Might just be a fluke or some other interference between time frames.

I suggest you to swap OPA2604 with OPA1612 or OPA1652 (if you really want FET-input opamp) and see if noise gets lowered. Of course, LM4562/LME49720 mioght be tested as well.

I'm not actually using this for any IEMs at the moment (they were a test), but I'll keep it in mind. LME49720 didn't work initially (I wrote about it in this post ), which is why I went with the OPA2134.

 

[DonH56]

I have not been following this thread... From the schematic @trl provided R9/R10 and R13/R14, or alternatively the two LEDs, are used to raise the 7x15 regulators' output voltages from the nominal fixed +/-15 V values to about +/-21 V. If you shorted R9 and R13 then you dropped the voltage to +/-15 V and reduced the output. That may be why the lower noise.

Again, I just happened to see this, so have not read all the posts and may be completely wrong.

 

[Sythrix]

I have not been following this thread... From the schematic @trl provided R9/R10 and R13/R14, or alternatively the two LEDs, are used to raise the 7x15 regulators' output voltages from the nominal fixed +/-15 V values to about +/-21 V. If you shorted R9 and R13 then you dropped the voltage to +/-15 V and reduced the output. That may be why the lower noise.

Again, I just happened to see this, so have not read all the posts and may be completely wrong.

R9 and R13 are not shorted. They now look like this, as the design called for when using the LEDs.

Since what I initially did (populated all spots instead of one of the 3 configurations) then there must have been something wrong with it, but I don't know at the moment.

 

[DonH56]

OK, that is fine, thanks. Measure the +/-15 V rails. You changed the voltage a little (up or down) when you pulled the R's, which may or may not be the root cause for the noise change...

I think you actually want to keep R10 and R14, however, to bias the LED. I think the choice is to use the LEDs D5/D6 or R9/R13 in the ground leg but R10/R14 should always be there. Without them, it is just the regulators' trickle bias (quiescent) current through the LEDs, which strikes me as not terribly stable.

 

[trl]

Search after "Configuration 2 - “LED reference”" in DIY Audio thread, from my understanding you might want to only use LEDs and smoothing caps, without the two resistors mounted. You should have 15V + 1.8V that drops on the LED, so about +/-16.8-16.9V. However, use a DMM and take some measurements; also, if you could borrow a scope it might be awesome, so you could test the amp with square-waves and check for pre-post ringing and if there's roll-off @ 20 Hz and @20 KHz.

For lower noise I suggest OPA1652; I've no idea why the LME49720 was not working...perhaps voltage rails were too high, so try LME49860 instead and see if it's working. Also OPA1612 can operate at +/-18V, so feel free to try it out too. Just take care of the power rails to stay within the +/-18V, otherwise you'll have trouble finding an opamp to work on such voltage.

Also, loose the 2 x 1uF caps from the audio input, see the AP measurements from DIY Audio and you'll see the bass roll-off. Just use 2.2uF or 4.7uF caps instead (better audio caps you can find, but don't spend more than 5 USD each).

L.E.: Leaving in place resistors and also LEDs would probably give you something like +/-17V; however, use a DMM to check the voltage, just to be sure you're using the opamps safely (LME49720 can operate at +/-15V).